From the Swedish Institute for Infectious Disease Control and the Department of Microbiology, Tumor and Cell Biology,

Karolinska Institutet, Stockholm, Sweden

EVOLUTION OF BETA-LACTAM RESISTANCE IN

KLEBSIELLA PNEUMONIAE

Sara Hæggman

Stockholm 2010

All previously published papers were reproduced with permission from the publisher.

Published by Karolinska Institutet. Printed by [name of printer]

© Sara Hæggman, 2010 ISBN 978-91-7409-938-6

ABSTRACT

K. pneumoniae is recognized as a common opportunistic pathogen. Numerous

reports have been published worldwide on outbreaks in different healthcare

settings. K. pneumoniae is inherently resistant to penicillins, including semi-

synthetic broad-spectrum penicillins. The drug of choice for empirical

treatment is often a cephalosporin. However, the use of cephalosporins is

known to select for extended-spectrum beta-lactamase (ESBL)-producing

strains.

The focus of this thesis is the beta-lactamase gene in K. pneumoniae, and its

relationship to beta-lactamase genes present in plasmids in gram-negative

bacteria. In Paper I, the intension was to identify presumed beta-lactamase

SHV-1-encoding plasmids in fecal Klebsiella isolates from neonates in Swedish

special care units. No such plasmids were detected, however. Instead, a

chromosomal beta-lactamase gene was identified in all K. pneumoniae, but in

none of the Klebsiella oxytoca isolates. This species-specific gene was seen in 10

allelic variants; some closely related to the prototypic plasmid-borne SHV-1

gene, indicating that an allelic variant of the K. pneumoniae chromosomal beta-

lactamase gene is the ancestor of the plasmid-borne SHV-encoding genes. In

Paper II, the observed diversity of the chromosomal K. pneumoniae beta-

lactamase gene was further investigated in order to study its evolution in

relation to the three phylogenetic groups of K. pneumoniae. Three sequence

groups, corresponding to the phylogenetic groups, were identified, blaSHV,

blaLEN, and blaOKP. In Paper III, the genetic context of blaSHV in K. pneumoniae

chromosomes and plasmids from various gram-negative bacteria was

analyzed. Plasmid-borne blaSHV genes were found to be surrounded by DNA

highly similar to the K. pneumoniae chromosome. IS26 elements flanked the

blaSHV regions. Nine distinct junctions between IS26 and K. pneumoniae

chromosomal DNA, and seven different region-lengths were identified. In

contrast to a high diversity observed among chromosomal sequences, only

two groups of plasmid sequences were seen.

This thesis has demonstrated that only one of three ancient K. pneumoniae

chromosomal beta-lactamase gene families, blaSHV, is found on plasmids. This

is possibly the result from a single IS26 mediated mobilization of blaSHV and

surrounding DNA from K. pneumoniae. The two groups of plasmid blaSHV

regions seen today could be the result of post-mobilization evolution

involving size reductions and nucleotide substitutions. We conclude that

mobilization of blaSHV from K. pneumoniae chromosomes is not a driving force

in the emergence of resistance in response to beta-lactam therapy. The spread

is more likely a consequence of mobilization of IS26 flanked blaSHV regions

between plasmids, and mobilization of plasmids between different bacteria.

LIST OF PUBLICATIONS

This thesis is based on the following papers, which are referred to in the text by their Roman numerals:

I. Hæggman, S., Löfdahl, S., and Burman, L.G. An allelic variant of the chromosomal gene for class A β-lactamase K2, specific for Klebsiella pneumoniae, is the ancestor of SHV-1 Antimicrobial Agents and Chemotherapy, 1997, 41(12): 2705-2709

II. Hæggman, S., Löfdahl, S., Paauw, A., Verhoef, J., and Brisse, S. Diversity and evolution of the class A chromosomal beta-lactamase gene in Klebsiella pneumoniae Antimicrobial Agents and Chemotherapy, 2004, 48(7): 2400-2408

III. Hæggman, S., and Löfdahl, S. Low initial transposition frequency of chromosomal blaSHV and subsequent evolution have formed the present population of acquired blaSHV

Manuscript

CONTENTS

1 Introduction ................................................................................................... 1

2 Klebsiella pneumoniae ................................................................................. 2

2.1 Physiology ......................................................................................... 2

2.2 Taxonomy ........................................................................................... 2

2.3 Healthcare-associated infections ........................................................ 3

2.4 Antibiotic treatment ............................................................................ 3

3 Beta-lactam antibiotics ................................................................................. 5

3.1 History ................................................................................................. 6

3.2 Mechanisms of action ......................................................................... 7

4 Beta-lactamases ............................................................................................ 8

4.1 Enzyme characteristics ....................................................................... 8

4.2 Classification....................................................................................... 9

4.3 Nomenclature .................................................................................... 10

4.4 Beta-lactamases in K. pneumoniae .................................................. 11

5 Identification of K. pneumoniae chromosomal beta-lactamase genes ...... 12

5.1 Paper I ............................................................................................... 12

6 Diversity of K. pneumoniae chromosomal beta-lactamase genes ............ 16

6.1 Paper II .............................................................................................. 16

7 Genetic contexts of blaSHV in K. pneumoniae chromosomes and

plasmids from different gram-negative bacteria ........................................ 20

7.1 Paper III ............................................................................................. 20

8 Acknowledgements .................................................................................... 26

9 References ................................................................................................... 27

LIST OF ABBREVIATIONS

bp Base pairs

ESBL Extended-spectrum beta-lactamase

PBP Penicillin-binding protein

PCR Polymerase chain reaction

1

1 INTRODUCTION

Antibiotic resistance is a serious problem in clinical medicine. For

example, the efficacy of treatment with the widely used beta-lactam

antibiotics is constantly challenged by the emergence of new resistant

bacterial strains.

Penicillin, the first industrially produced beta-lactam antibiotic, has been

in clinical use since the 1940s. Soon after its introduction into clinical

praxis, however, resistant bacterial strains emerged. New chemically

modified beta-lactam antibiotics were therefore successively developed

by the pharmaceutical companies. The broad-spectrum penicillins were

soon followed by a large number of cephalosporins. Cefotaxime, the

first so called “third generation” cephalosporin was introduced in the

early 1980s and it is still one of the most widely used cephalosporins.

In gram-negative bacteria, beta-lactamase production is considered the

main antibiotic resistance mechanism. Beta-lactamases are enzymes that

inactivate beta-lactam antibiotics. This group of enzymes comprises

many variants with different spectra of activity. New variants are

continually being identified around the world, the most worrisome

being the extended-spectrum beta-lactamases (ESBLs) belonging to the

enzyme families TEM, SHV, and CTX-M.

Genes encoding beta-lactamases are found both in bacterial

chromosomes and plasmids. Since the 1980s there have been numerous

reports from different healthcare settings worldwide on outbreaks

caused by ESBL-producing pathogenic gram-negative bacteria. Many of

these reports involve Klebsiella pneumoniae, which is recognized as one of

the most common causes of healthcare associated bacterial infections.

The subject of this thesis is beta-lactam resistance in K. pneumoniae, with

focus on the inherent chromosomal beta-lactamase gene. I have

investigated its diversity, evolution, and genetic context. I have also

compared it to plasmid-borne SHV genes in order to study evolutionary

relationships between chromosomal and plasmid-borne SHV beta-

lactamase genes, and to clarify mechanisms involved in the worldwide

dissemination of different SHV ESBL-encoding genes.

2

2 KLEBSIELLA PNEUMONIAE

2.1 PHYSIOLOGY

Klebsiella pneumoniae is a non-motile gram-negative rod-shaped bacterium,

≤ 6 µm long and ≤ 1 µm in diameter, which can grow both aerobically

and anaerobically (Brisse 2006). The cells are often surrounded by a

polysaccharide capsule, which prevents phagocytosis. The capsule is

regarded as a virulence determinant, and most clinical isolates are

capsulated (Favre 1999). There are 77 capsular (K) serotypes described,

and some of them have been associated with severe infections in

humans and animals (Brisse, Fevre et al. 2009). K. pneumoniae is a

common member of the human intestinal flora, and it is said to be

ubiquitous, meaning that it can be found almost everywhere, for

example also in water, soil, and plants (Podschun, Pietsch et al. 2001),

(Brisse and Duijkeren 2005). Some strains isolated from plants are

nitrogen-fixing and of interest since they can increase plant growth

under agricultural conditions. One of the three publically available K.

pneumoniae whole genome sequences is that of a nitrogen-fixing strain

isolated from the interior of nitrogen-efficient maize plants (Fouts, Tyler

et al. 2008).

2.2 TAXONOMY

K. pneumoniae belongs to the family Enterobacteriaceae (Francino 2006);

(Grimont PAD 2005; Brisse 2006). It is the type species of the genus

Klebsiella, which was named in honor of the German microbiologist

Edwin Klebs who lived 1834-1913 (Trevisan 1885). The first Klebsiella

strain ever described was a capsulated bacillus isolated from a patient

with rhinoscleroma (Frisch 1882).

The type strain of K. pneumoniae is ATCC 13883 (NCTC 9633, CDC

298-53, CIP 82.91). This strain belongs to one of three subspecies,

namely pneumoniae. The other two K. pneumoniae subspecies are ozaenae

and rhinoscleromatis. The definitions of the subspecies are not based on

genomic distinctness but on pathogenesis criteria (Brisse 2006).

The nomenclature for organisms within the genus Klebsiella has been

confusing. For example, the existence of an indole positive species, now

known as Klebsiella oxytoca, was questioned. It was then regarded as a

3

biogroup of K. pneumoniae (Edwards and Ewing 1972), (Ørskov 1974).

By DNA relatedness studies, however, it was clarified that K. oxytoca is

distinct from K. pneumoniae at the species level (Jain 1974), (Brenner

1977).

When the genetic diversity of the species K. pneumoniae was investigated,

three sequence clusters, or phylogenetic groups, were identified (Brisse

and Verhoef 2001). The groups were named KpI, KpII, and KpIII.

Recently, it has been shown that even more phylogenetic groups exist

within K. pneumoniae (Jonas, Spitzmuller et al. 2004).

Most K. pneumoniae infections are caused by strains belonging to the

phylogenetic group KpI (Brisse 2004).

2.3 HEALTHCARE-ASSOCIATED INFECTIONS

K. pneumoniae is recognized as a common opportunistic pathogen. It

accounts for a significant proportion of healthcare-associated, or

nosocomial, infections that are frequently caused by gram-negative

enterobacteria (Podschun and Ullman 1998a). In many studies, it is one

of the three most common gram-negative pathogens, together with E.

coli and Pseudomonas aeruginosa (Richards 2000), (Garcia de la Torre,

Romero-Vivas et al. 1985), (Williams and Thomas 1990). Age is one of

the predisposing factors, i.e., very young or very old (Feldman 1990).

The reservoir for the K. pneumoniae strains is often the intestinal tracts of

the patients.

Numerous reports have been published worldwide on outbreaks caused

by K. pneumoniae in different healthcare settings, like neonatal wards,

nursing homes, and intensive care units (Liu, Gur et al. 1992), (Arpin,

Dubois et al. 2003), (Gniadkowski, Palucha et al. 1998), (Livermore and

Yuan 1996), (Babini and Livermore 2000), (Sadowski and al. 1979),

(Lytsy, Sandegren et al. 2008).

2.4 ANTIBIOTIC TREATMENT

K. pneumoniae is inherently resistant to penicillins, including semi-

synthetic broad-spectrum penicillins. Therefore, the drug of choice for

empirical treatment is often a cephalosporin. However, the use of

cephalosporins is known to select for resistant K. pneumoniae strains

4

(Livermore 1991), (Bedenic 2002). This is of great concern in human

healthcare around the world. The number of K. pneumoniae strains

producing ESBL variants of the widespread plasmid-encoded beta-

lactamases belonging to the enzyme families TEM, SHV, and CTX-M

are constantly increasing (Bradford 2001), (Jacoby and Munoz-Price

2005), (Paterson, Hujer et al. 2003), (Paterson and McCormack 2003),

(Steward, Rasheed et al. 2001), (Winokur, Canton et al. 2001).

3 BETA-LACTAM ANTIBIOTICS

All beta-lactam antibiotics contain a beta

(Rolinson and Geddes 2007)

beta-lactamases

inactivate the antibiotic

Fig. 1. The core structure of penicillins (left) and cephalosporins (right). The four

the beta-lactam ring. The five

ring in cephalosporins is a dihydrothiazine ring.

penicillins, e.g., a benzyl group in penicillin G

in different cephalosporins.

The four major grou

cephalosporins, carbapenems, and

lactamase gene in focus of this thesis exists in variants which encode

enzymes that mainly inactivate penicillins and cephalosporins.

Table 1. Examples of beta

Group

Penicillins

Cephalosporins

Carbapenems

Monobactams

LACTAM ANTIBIOTICS

lactam antibiotics contain a beta-lactam ring, hence the name

(Rolinson and Geddes 2007). This molecular structure is the target for

lactamases, which by hydrolysis open the ring and thereby

inactivate the antibiotic (Fig. 1).

The core structure of penicillins (left) and cephalosporins (right). The four

lactam ring. The five-membered ring in penicillins is a thiazolidine ring. The six

ring in cephalosporins is a dihydrothiazine ring. The R indicates various side groups in different

, e.g., a benzyl group in penicillin G. R1 and R

2 indicate positions for various

in different cephalosporins.

The four major groups of beta-lactam antibiotics are penicillins,

cephalosporins, carbapenems, and monobactams (Table 1).

lactamase gene in focus of this thesis exists in variants which encode

enzymes that mainly inactivate penicillins and cephalosporins.

of beta-lactam antibiotics in the four major groups

Beta-lactam antibiotic(s) Commonly referred to as

Penicillin G, penicillin V

Ampicillin, piperacillin

Broad-spectrum penicillins

Cephalothin

Cefuroxime

Cefotaxime, ceftazidime

First generation cephalosporins

Second generation cephalosporins

Third generation cephalosporins

Imipenem, meropenem

Aztreonam

5

lactam ring, hence the name

. This molecular structure is the target for

, which by hydrolysis open the ring and thereby

The core structure of penicillins (left) and cephalosporins (right). The four-membered ring is

membered ring in penicillins is a thiazolidine ring. The six-membered

cates various side groups in different

positions for various side groups

lactam antibiotics are penicillins,

Table 1). The beta-

lactamase gene in focus of this thesis exists in variants which encode

enzymes that mainly inactivate penicillins and cephalosporins.

Commonly referred to as

spectrum penicillins

First generation cephalosporins

Second generation cephalosporins

Third generation cephalosporins

6

3.1 HISTORY

Beta-lactam antibiotics have been used as therapeutic agents since the

1940s. They are excellent drugs because they are non-toxic and well

tolerated by most patients.

Less than one hundred years ago, in the 1920s, Sir Alexander Fleming

made the ground-breaking discovery that a certain Penicillium mould

produced a powerful antibacterial substance (Flemming 1929). He

named the filterable active agent penicillin, and reported that the action

of penicillin was marked on the pyogenic cocci and the diphtheria group

of bacilli, and that for example the coli-typhoid bacteria were quite

insensitive. Fleming, however, was not able to purify penicillin. It was

not until the late 1930s that this was accomplished (Chain 1938).

Penicillin G (benzylpenicillin) was the first beta-lactam antibiotic in

clinical use (Florey and Florey 1943). A major drawback of this drug is

that it cannot be administered orally due to its lack of stability in the

acid stomach. In the 1950s, penicillin V (phenoxymethylpenicillin), was

developed (Brandl, Giovannini et al. 1953). This semi-synthetic

derivative is acid stable. Both penicillin G and V, however, have rather

limited spectrum of activity, and are not suitable for treating infections

caused by gram-negative bacteria.

Ampicillin is a broad-spectrum penicillin that was developed in the

1960s by chemical modification of the side chain of the beta-lactam ring

of benzylpenicillin (Rolinson and Stevens 1961). Its spectrum of activity

includes gram-negatives like E. coli. Ampicillin has been followed by

many other beta-lactam antibiotics, which have been developed in order

to further increase the spectrum of activity.

A wide range of semi-synthetic beta-lactam antibiotics have been

developed by pharmaceutical companies in response to the emergence

of resistant bacterial strains, which were soon selected for by the

therapeutic use of every new beta-lactam antibiotic (Livermore 2009).

Cephalosporins are beta-lactam antibiotics originally isolated from the

mould Cephalosporium (Murray, Rosenthal et al. 2009). They differ from

penicillins in having a dihydrothiazine ring fused with the beta-lactam

ring (Fig. 1). This gives more opportunities for biochemical

modifications (at positions R1 and R2), in order to expand the spectrum

7

of activity and improve the pharmacokinetic properties of the drug.

Cephalosporins, in general, have enhanced activity against gram-

negative bacteria. Commonly used semi-synthetic cephalosporins are

cefuroxime, cefotaxime, and ceftazidime (Knothe and Dette 1983).

3.2 MECHANISMS OF ACTION

Beta-lactam antibiotics only kill growing bacteria. They bind so called

penicillin binding proteins (PBPs), which are enzymes involved in cell

wall synthesis (Sauvage, Kerff et al. 2008). The PBPs are located on the

outer side of the cytoplasmic membrane. In gram-negative bacteria this

is in the periplasm. Some PBPs are transpeptidases that catalyze the

cross-linking of glycan strands in the nascent peptidoglycan. When these

enzymes are inactivated cell wall synthesis becomes severely disturbed,

which leads to cell lysis.

The mechanism of action is explained by the structural similarity

between the beta-lactam ring and the peptidoglycan building block acyl-

D-alanyl-D-alanine (Tipper and Strominger 1965). A covalent bond

formed between the beta-lactam ring and an active site serine residue in

the PBP results in inactivation of the PBP.

It is the fact that peptidoglycan, the major structural component of

most bacterial cell walls, is unique and essential for bacteria that makes

beta-lactam antibiotics excellent non-toxic drugs for humans.

8

4 BETA-LACTAMASES

Resistance to beta-lactam antibiotics can have different causes (Table 2).

Table 2. The three major beta-lactam resistance mechanisms

Mechanism Effect on the antibiotic Reference

Loss of outer membrane proteins Prevented from reaching its site of action (Nikaido 2003)

Altered PBPs Prevented from binding target enzyme (Spratt 1994)

Beta-lactamase production Inactivated irreversibly (Jacoby and

Munoz-Price

2005)

Production of one or more beta-lactamases is by far the most common

mechanism for beta-lactam resistance among gram-negative bacteria

(Livermore 2009). K. pneumoniae strains, for example, often carry

plasmids producing one or more beta-lactamase variants (Pitout and

Laupland 2008).

4.1 ENZYME CHARACTERISTICS

Beta-lactamases hydrolyze the amide bond of the beta-lactam ring

(Ghuysen 1991; Livermore 2009). The molecular mass of these enzymes

is ~30 kDa (~280 amino acid residues). In gram-negative bacteria, beta-

lactamases are found as soluble proteins in the periplasm. Beta-

lactamases and PBPs are structurally related and share certain

mechanistic features (Massova and Mobashery 1998). These enzymes

probably have a common origin (Massova and Mobashery 1999;

Gniadkowski 2008).

Some beta-lactamases are zinc-dependent enzymes. Metal-independent

beta-lactamases are, however, more common. They contain an active

site serine residue to which the antibiotic is covalently bound, via an

ester bond, as a catalytic intermediate. The ester bond is subsequently

hydrolyzed and the inactivated antibiotic is released from the enzyme.

The beta-lactamase is then ready for a new catalytic cycle. This

mechanism is analogous to the binding of beta-lactam antibiotics to

PBPs. The main difference is that the covalent bond formed between

9

the antibiotic and the active site serine in the PBP is not, or very slowly,

hydrolyzed. Therefore, the enzyme activity of the PBP is blocked.

Beta-lactamase activity can be demonstrated by using, for example,

nitrocefin, a chromogenic cephalosporin reagent (Galleni and Frère

2007).

Beta-lactamase activity can be inhibited by clavulanic acid, a beta-lactam

compound that was discovered in 1977 (Reading and Cole 1977). This

compound and other beta-lactamase inhibitors (sulbactam and

tazobactam) are used in combination with beta-lactam antibiotics as as

therapeutic agents. For example, the drug Augmentin contains a mixture

of amoxicillin and clavulanic acid.

4.2 CLASSIFICATION

A large number of beta-lactamase variants have been identified since the

report in 1940 on an E. coli enzyme able to destroy penicillin (Abraham

and Chain 1940). Different classification schemes for these enzymes

have been presented. The most recent scheme was published this year

(Bush and Jacoby 2010). The first attempts to classify beta-lactamases

were based on functional and biochemical characteristics of the enzyme,

like, substrate profile and isoelectric point. Later, amino acid sequence

information was added to the schemes (Bush, Jacoby et al. 1995).

The latest classification schemes for beta-lactamases include four

molecular classes, A, B, C, and D, based on amino acid sequence

information. They also include four functional groups, 1 to 4, which are

based on hydrolytic and inhibition properties of the enzymes. The

inhibitors used in the latest classification scheme are clavulanic acid,

tazobactam and EDTA.

Molecular class A, C, and D comprise the serine beta-lactamases, and

class B the zinc-dependent metalloenzymes. Class A and D include

penicillinases and cephalosporinases, most of which are inhibited by

clavulanic acid or tazobactam. Class C include cephalosporinases that

are not, or poorly, inhibited by clavulanic acid or tazobactam. Class B

comprises the metalloenzymes, which are inhibited by EDTA and have

carbapenems as distinctive substrates. Representative enzymes of the

different molecular classes are presented in Table 3.

10

Table 3. Classification of common beta-lactamases

Molecular class Functional group Beta-lactamase(s)

C 1 AmpC (E. coli)

A 2b

2be

2br

2ber

TEM-1, TEM-2, SHV-1

TEM-3, SHV-2, CTX-M-15

TEM-30, SHV-10

TEM-50

D 2d

2de

2df

2f

OXA-1

OXA-11

OXA-23

KPC-2

B 3a IMP-1, VIM-1

4.3 NOMENCLATURE

The nomenclature of beta-lactamases is comprehensively covered in a

recent minireview (Jacoby 2006). The enzymes have been named after,

for example, biochemical properties, the strain producing it, the person

first characterizing it, or the patient providing the first sample. Table 4

presents examples of beta-lactamases found in K. pneumoniae.

Table 4. Some beta-lactamases found in K. pneumoniae

Beta-lactamase Derivation of name Reference

SHV Sulfhydryl reagent variable (Matthew,

Hedges et al.

1979)

LEN From K. pneumoniae strain LEN-1 (Arakawa, Ohta

et al. 1986)

OKP Other K. pneumoniae beta-lactamase (Paper II)

TEM Named after the patient (Temoneira) providing the first

sample

(Datta and

Kontomichalou

1965)

CTX-M Active on cefotaxime, first isolated in Munich (Bauernfeind,

Grimm et al.

1990)

KPC K. pneumoniae carbapenemase (Yigit, Queenan

et al. 2001)

11

Some of the enzymes show many closely related variants, which form

enzyme families. Such families are TEM, SHV, and CTX-M. At present

there are for example more than 130 SHV variants described. See

publically available databases for updated information

(http://www.lahey.org/Studies/; http://www.pasteur.fr/recherche/

genopole/PF8/betalact_en.html).

For clarity, when mentioning an ESBL, it is suggested to always include

the enzyme family name, for example SHV ESBL (Livermore 2008).

4.4 BETA-LACTAMASES IN K. PNEUMONIAE

K. pneumoniae is inherently resistant to penicillins and early

cephalosporins due to constitutive production of a chromosomally

encoded class A group 2b beta-lactamase (Petit, Ben-Yaghlane-

Bouslama et al. 1992), (Paper I). In addition to this enzyme, many K.

pneumoniae strains produce one or more plasmid-mediated beta-

lactamases. The most common belong to the enzyme families TEM,

SHV, and CTX-M (Jacoby 1997), (Gniadkowski 2008), (Hawkey 2008),

(Elhani, Bakir et al. 2010). SHV ESBLs have been demonstrated in K.

pneumoniae since 1983 (Knothe, Shah et al. 1983), (Kliebe, Nies et al.

1985). Today, the most common SHV ESBLs worldwide are SHV-2,

SHV-5 and SHV-12 (Hrabak, Empel et al. 2009). New SHV variants

still emerge (Jones, Tuckman et al. 2009). In recent years, reports on

plasmid-mediated enzymes belonging to the CTX-M family have

become more and more frequent (Livermore, Canton et al. 2007).

12

5 IDENTIFICATION OF K. PNEUMONIAE

CHROMOSOMAL BETA-LACTAMASE GENES

5.1 PAPER I

An allelic variant of the chromosomal gene for class A beta-

lactamase K2, specific for Klebsiella pneumoniae, is the ancestor

of SHV-1

The Klebsiella strains used in this study were 172 fecal non-duplicate

isolates from neonates in 22 Swedish special care units. These strains

had been included in two previous studies; one focusing on the

epidemiology of strains of different Enterobacteriaceae species (Tullus,

Berglund et al. 1988), and one focusing on the epidemiology of the

plasmid-mediated beta-lactamases TEM-1, OXA-.1, and SHV-1

(Burman, Haeggman et al. 1992). Also 15 international reference strains,

including the type strains of K. pneumoniae subsp. pneumoniae, K.

pneumoniae subsp. ozaenae, K. pneumoniae subsp. rhinoscleromatis, Klebsiella

oxytoca, K. planticola, and K. terrigena were included in the study.

The aim of the study was to identify presumed SHV-1-encoding

plasmids among the fecal Klebsiella isolates and to find out whether

SHV-1-encoding genes were carried by promiscuous plasmids or not.

However, in Southern blot analyses of plasmid DNA preparations, we

failed to detect any hybridization between plasmid bands and the SHV-1

probe used. This probe, a gel purified 352 bp PvuII intragenic fragment

obtained by digestion of plasmid pMON38, was the one used in the

previous colony hybridization assay (Burman, Haeggman et al. 1992).

The recombinant plasmid pMON38 is derived from the SHV-1-

encoding plasmid R974 (Mercier and Levesque 1990).

To increase the sensitivity of our Southern blot hybridizations, we tested

another type of probe – a PCR amplicon. The PCR primers designed

for this purpose were based on the sequence of the intragenic 352 bp

fragment of SHV-1R974 present in pMON38, which was used as PCR

template. This PCR yielded a 231 bp amplicon which was labeled with

[α-32P]dCTP and used as probe in subsequent Southern blot analyses.

Also this probe failed to hybridize with plasmid DNA. By using the

PCR-derived probe the hybridization signal was increased and the

background was decreased. However, weak hybridization signal was

seen with chromosomal DNA present in low concentration in some of

13

the plasmid preparations. This finding prompted us to perform

Southern blot analysis of genomic DNA. By doing this, we detected

hybridization with chromosomal DNA from all tested K. pneumoniae

strains – 20 fecal isolates and three reference strains (ATCC 13883T,

1976E, and LEN-1).

The developed SHV-1-based PCR was used to screen 187 Klebsiella

strains. This revealed the presence of an SHV-1 or SHV-1-related beta-

lactamase gene in all 116 K. pneumoniae strains included in the study, and

the lack of such a gene in all 69 Klebsiella oxytoca strains as well as in the

K. planticola and K. terrigena type strains.

The finding in K. pneumoniae of what appeared to be a species-specific

beta-lactamase gene, closely related to the plasmid-borne SHV-1 beta-

lactamase gene, was rather unexpected. At that time SHV-1 was

generally regarded as a plasmid-mediated beta-lactamase (Bush, Jacoby

et al. 1995), (Fuster, Roy et al. 1993) even though some reports had

indicated that occasional K. pneumoniae strains could produce

chromosomally encoded SHV-1-like enzymes (Matthew and Harris

1976), (Nugent and Hedges 1979), (Petit, Ben-Yaghlane-Bouslama et al.

1992). The only K. pneumoniae chromosomal beta-lactamase gene that

had been identified and sequenced at that time was the one encoding

beta-lactamase LEN-1 (Arakawa, Ohta et al. 1986). Both SHV and LEN

enzymes are distinct from AmpC, i.e., the chromosomal beta-lactamase

produced by E. coli and many other species of Enterobacteriaceae.

Fig. 2. Restriction fragment length polymorphism analysis of DNA from five fecal K. pneumoniae isolates. Agarose gel of total bacterial DNA preparations digested with BglII (left), and the corresponding autoradiogram after Southern blot hybridization (right). Undigested DNA preparations of K. pneumoniae 1976E (C1) and E. coli J53-2 containing pMON38 (C2). The position of the blaSHV-positive ~5-kb BglII fragment is indicated by an arrow. (Adapted from Paper I).

14

Further characterization of K. pneumoniae genomic DNA, using

restriction fragment length polymorphism analysis, showed that the

beta-lactamase gene was located within a conserved chromosomal

region. The SHV-1 PCR probe hybridized to a ~5 kb BglII fragment in

all tested strains (Fig. 2).

The negative results of conjugation experiments performed were also in

support of the notion of a chromosomal location of the SHV-1-like

beta-lactamase gene in K. pneumoniae. Transfer of ampicillin resistance

was only demonstrated for one of the tested isolates. This was the only

K. pneumoniae isolate that had been colony hybridization positive for

both SHV-1 and TEM-1.

Analytical isoelectric focusing of chromosomally encoded K. pneumoniae

beta-lactamases identified two major groups. Enzymes with isoelectric

point (pI) 7.6 were the most common. They were produced by isolates

collected from neonates in 18 of the 22 special care units. Enzymes

focusing at pI 7.1 were produced by isolates from 8 special care units

only. A pI of 7.6 is characteristic for SHV-1 beta-lactamase and pI 7.1 is

characteristic for LEN-1.Other pIs were also detected (for details see

Table 1, Paper I).

Randomly selected SHV-1-PCR amplicons were subjected to DNA

sequencing. Alignment of sequences and tree analysis demonstrated the

same groupings as seen by isoelectric focusing. Sequences from strains

producing a pI 7.6 enzyme formed one cluster, as did the strains

producing pI 7.1 enzymes (Fig. 4, Paper I).

Minimal inhibitory concentrations (MICs) were determined for

ampicillin, ampicillin in the presence of clavulanic acid, piperacillin,

cephalothin, cefotaxime, and aztreonam. All strains exhibited broad-

spectrum beta-lactamase activity, and the beta-lactamase inhibitor

clavulaninc acid markedly lowered the ampicillin MICs. There were no

distinctions between the K. pneumoniae isolates producing SHV-1-like

beta-lactamase and the ones producing LEN-1-like enzymes. No

extended-spectrum beta-lactamase activity was detected.

In summary, what started as a search for plasmids carrying SHV-1 beta-

lactamase genes resulted in the identification of a chromosomal beta-

lactamase gene present in all K. pneumoniae, i.e., a species-specific beta-

lactamase gene. In our material, this gene was seen in 10 allelic variants.

15

Most of the variants were closely related to the prototypic plasmid-

borne SHV-1 beta-lactamase gene. This made us propose that an allelic

variant of the K. pneumoniae chromosomal beta-lactamase gene is the

ancestor of the plasmid-borne SHV-encoding genes observed frequently

in K. pneumoniae and other pathogenic gram-negative bacteria.

16

6 DIVERSITY OF K. PNEUMONIAE

CHROMOSOMAL BETA-LACTAMASE GENES

6.1 PAPER II

Diversity and evolution of the class A chromosomal beta-

lactamase gene in Klebsiella pneumoniae

Paper II reports extended analysis of the evolutionary relationships

between different chromosomal beta-lactamase gene variants. The aim

of this study was to investigate whether beta-lactamase diversification

has occurred as part of a natural, long-term evolutionary process, or

whether the presence of a chromosomal beta-lactamase gene in K.

pneumoniae is the result of a recent horizontal acquisition followed by

clonal expansion in response to selective pressure of various beta-lactam

antibiotics used in clinical medicine.

Twenty K. pneumoniae strains were randomly selected from the two

major groups identified in our previous study (Paper I); ten of them

produced a pI 7.6 SHV beta-lactamase and the other ten produced a pI

7.1 LEN beta-lactamase. One of the two strains found to have a beta-

lactamase gene different from blaSHV and blaLEN was also included. The

excluded strain was negative in Southern blot analysis. In addition to

these 21 strains, we included seven strains from a previous taxomomic

study of K. pneumoniae, in which three K. pneumoniae phylogenetic groups,

KpI, KpII, and KpIII, were identified (Brisse et al. 2001).

We studied the nucleotide diversity and evolution of the K. pneumoniae

chromosomal beta-lactamase gene and two housekeeping genes; gyrA,

coding for subunit A of gyrase, and mdh, coding for malate

dehydrogenase. The gyrA gene had previously been used as a marker for

the three K. pneumoniae phylogenetic groups (Brisse et al. 2001).

A 789-bp portion of the beta-lactamase genes was sequenced from the

28 strains included in this study, i.e., a larger part than in our previous

study (Paper I) in which we sequenced 231-bp PCR amplicon. The

obtained sequences formed three distinct groups. This corresponded

fully to our earlier results (Paper I). One of the groups comprised

sequences from all strains producing pI 7.6 SHV beta-lactamases and

two KpI reference strains. A second group comprised sequences from

all strains producing pI 7.1 LEN beta-lactamases and the two KpIII

17

reference strains. The third group comprised sequences from strains

producing beta-lactamases different from SHV and LEN, both as

determined by isoelectric focusing and by nucleotide sequencing. Two

of these strains were the two KpII reference strains. None of the

sequences in the third group matched closely to any sequence in public

databases. Even though this group was more heterogeneous than the

two other groups, we decided to assign a new beta-lactamase gene

family and named it blaOKP (other K. pneumoniae beta-lactamases).

The correspondence between chromosomal beta-lactamase gene

variants and K. pneumoniae phylogenetic groups is illustrated in Fig. 3.

The data indicate parallel evolution of bla and gyrA. All sequences

belonging to the blaSHV group were from KpI strains, as determined by

gyrA sequencing, all blaLEN sequences were from KpIII strains, and the

blaOKP sequences were all from KpII strains.

Fig. X. Phylogenetic trees of bla (left) and gyrA (right) sequences from

28 K. pneumoniae strains (adapted from Paper II, figures 1 and 2).

By analyzing thirty-four additional K. pneumoniae strains (previously

assigned to any of the three phylogenetic groups KpI, KpII, and KpIII)

by PCRs newly developed in order to specifically amplify blaSHV,

blaLEN, and blaOKP, respectively, we fully confirmed the

correspondence, seen by nucleotide sequencing, between chromosomal

beta-lactamase genes and phylogenetic groups.

Fig. 3. Phylogenies of the K.pneumoniae chromosomal beta-lactamase gene (left) and the gyrA gene (right). The tree was obtained by the neighbor-joining method. The root was determined using the DNA sequence for beta-lactamase TEM-1, the closest known relative to the chromosomal beta-lactamase of K. pneumoniae. Because of the long branch leading to the TEM-1 sequence, the root position is represented by a triangle in order to increase the scale of the figure so that the relationships among K. pneumoniae sequences are clearly visible.

The results supported the high diversity among the K. pneumoniae

chromosomal beta-lactamase genes seen previously (Paper I). Alignment

of the 789-bp sequences from the 28 strains revealed 156 polymorphic

18

sites. In total, we identified 24 distinct bla alleles and 17 deduced amino

acid sequences, including one new SHV beta-lactamase variant, seven

new LEN variants, and four OKP variants. Interestingly, both strains

isolated from plants produced LEN beta-lactamases. This is interesting

since the beta-lactamase gene in the single full genome sequence

publically available presently from a plant isolate is a blaLEN gene (Fouts

et al. 2008). However, more plant isolates need to be analyzed before

any conclusions about the distribution of different beta-lactamase gene

families among different K. pneumoniae hosts one could be drawn.

In addition to nucleotide sequencing, the 28 K. pneumoniae strains were

subjected to MIC determinations. All strains were found to produce a

typical class A group 2 broad-spectrum penicillinase inhibited by

clavulanic acid, according to the Bush-Jacoby-Medeiros classification

scheme (Bush et al. 1995). No ESBL activity was detected. This was in

agreement with the deduced amino acid sequences, which did not reveal

any substitutions associated with extended-spectrum beta-lactamase

activity (Hujer et al. 2001).

The strikingly high correlation demonstrated between chromosomal

beta-lactamase gene variants and K. pneumoniae phylogenetic groups

encouraged us to investigate whether the beta-lactamase gene had been

inherent to this species for a long time. In order to do this, we estimated

the time since the divergence of the K. pneumoniae phylogenetic groups.

This was done by analyzing the mdh gene sequences based on the

molecular clock hypothesis, appreciating that DNA sequence variation

cannot be perfectly explained this way. The molecular clock hypothesis

states that the evolutionary rate of a gene is roughly constant among

different lineages. Under the molecular clock hypothesis, sequence

variation at synonymous sites is used to estimate the time since the

existence of the last common ancestor. The reasons for choosing the

mdh gene were that (i) the rate of evolution for this gene could be

calibrated because nucleotide sequences of this gene were previously

determined for natural populations of E. coli and Salmonella enterica, and

(ii) the time since divergence between E. coli and S. enterica was

previously estimated by different approaches. By using the extreme

values of the various estimates of the time since divergence between E.

coli and S. enterica, 30 and 140 million years, to calibrate the rate of

substitution, we arrived at an estimated time since divergence between

19

the two most distantly related phylogenetic groups, KpI (SHV-

producing strains) and KpII (LEN-producing strains), of 6 to 28

million years.

In summary, our results show that the chromosomal beta-lactamase

gene has not been acquired recently by K. pneumoniae in response to

clinical use of beta-lactam antibiotics. It has rather evolved along K.

pneumoniae phylogenetic lineages for millions of years – blaSHV along the

KpI lineage, blaLEN along the KpIII lineage, and blaOKP along the KpII

lineage. The hypothesis that clinical use of beta-lactam antibiotics would

have selected for acquisition of this gene is also challenged by the

finding of the same gene variant, blaLEN, in both plant isolates and

human clinical isolates.

20

7 GENETIC CONTEXTS OF BLASHV IN K.

PNEUMONIAE CHROMOSOMES AND

PLASMIDS FROM DIFFERENT GRAM-

NEGATIVE BACTERIA

7.1 PAPER III

Low initial transposition frequency of chromosomal blaSHV and

subsequent evolution have formed the present population of

acquired blaSHV

Paper III started as a study of the genetic context of a K. pneumoniae

chromosomal SHV-1 beta-lactamase gene, blaSHV-1. This was done in

order to investigate whether blaSHV-1 was located within a ≥5 kb

transposable element as indicated by previous restriction length

polymorphism analysis (Paper I, Fig. 2).

At the time, 1998, blaSHV was generally regarded as a plasmid-borne

gene. Evidence that the SHV-1 beta-lactamase gene existed as part of a

transposon of molecular mass 9.5 megadaltons (~14.3 kb) in two

unrelated plasmids had been published (Nugent and Hedges 1979).

Results in Papers I and II, however, supported the notion of the

presence of an inherent chromosomal beta-lactamase gene in K.

pneumoniae, belonging to any of the three gene families blaSHV, blaLEN, and

blaOKP which are specific for the three phylogenetic groups of K.

pneumoniae (Paper II, Figs. 1 and 2).

K. pneumoniae ATCC 13883T was chosen for this part of the study. The

reasons for this were that (i) this is the type strain of K. pneumoniae, i.e.,

an “old” strain, isolated in the pre-antibiotic era, which has not been

under selective pressure of clinical use of beta-lactam antibiotics, and (ii)

the nucleotide sequence of the chromosomal blaSHV-1 gene in this strain

is highly similar to the sequence of the prototype blaSHV-1 in plasmid

R974 (Paper I, Fig. 3).

This study started when the number of publically available nucleotide

sequences was very low compared to today. The only publically

available chromosomal K. pneumoniae beta-lactamase gene sequence was

that of blaLEN-1 (Arakawa, Ohta et al. 1986). This meant that the

chromosomal blaSHV-1 and surrounding DNA had to be cloned from K.

21

pneumoniae ATCC 13883T before sequencing could be performed. The

blaSHV-1 gene was located by Southern blot hybridization to an 8.4 kb

EcoRI fragment, which was cloned into E. coli and selected for by using

a cloning vector containing a kanamycin resistance gene. Transformants

resistant to both ampicillin and kanamycin were selected and screened

by blaSHV PCR. A shotgun library was constructed from one blaSHV

positive recombinant plasmid. The 8.4 kb EcoRI insert was agarose gel

purified and randomly fragmented. The fragments were modified at

their ends with T4 and Taq polymerase and then cloned by using the

cloning vector pGEM-T Easy. Clones containing inserted K. pneumoniae

DNA were selected on agar plates containing ampicillin and IPTG/X-

Gal for blue/white screening. Selected clones were sequenced using

vector specific primers. After having completed the 8.4 kb sequence and

not detected any genes with known function or any mobile genetic

elements, the sequence was extended beyond the EcoRI site upstream of

blaSHV-1. This was done by inverse PCR and primer walking. The final

10.6 kb sequence contained 10 open reading frames, of which only three

represented genes of known function, namely blaSHV-1, lacY, and lacZ

(Fig. 4).

Fig. 4. Schematic representation of the genetic context of the chromosomal blaSHV-1 in

K. pneumoniae ATCC 13883T (adapted from Paper III, Fig. 1).

In april 2010, there were three K. pneumoniae complete genome

sequences publically available: GenBank accession nos. CP000647 (a

clinical isolate carrying blaSHV), CP000964 (a nitrogen-fixing plant

endophyte carrying blaLEN), and AP006725 (a clinical isolate carrying

blaSHV). In all of them the chromosomal beta-lactamase gene was located

in the same genetic context as blaSHV-1 in K. pneumoniae ATCC 13883T.

The GC content, 57%, of these chromosomes was similar in our 10.6

kb sequence, 60%. This supported the notion that the chromosomal

beta-lactamase gene is inherent to K. pneumoniae and not part of a

horizontally acquired transposon or other type of mobile genetic

element.

lacZ lacY unknown blaSHV-1 ygbI ygbJ ygbK ygbL ygbM ygbN

EcoRI EcoRIBglIIBglIIBglII

1000 2000 3000 4000 5000 70006000 8000 9000 10000

22

By comparing the identified chromosomal 10.6 kb sequence of K.

pneumoniae ATCC 13883T to plasmid sequences publically available at the

time (late 1990s), we found plasmid-located blaSHV to be surrounded by

DNA highly similar to our K. pneumoniae chromosomal DNA (REFS?).

Therefore, in order to gain further knowledge about evolutionary

relationships between chromosome- and plasmid-encoded SHV beta-

lactamases, we investigated the genetic context of blaSHV also in

plasmids. This was started by performing re-sequencing of blaSHV-1 and

surrounding DNA from an E. coli strain (BAB273) producing a pI 7.6

SHV-1 beta-lactamase. We chose to work with E. coli because the

chromosomal beta-lactamase gene inherent to this species, blaAmpC, is

different from blaSHV and would not cause ambiguous sequencing

results. With the 10.6 kb K. pneumoniae ATCC 13883T sequence available

it was possible to design primers for direct sequencing of plasmid DNA.

By primer walking a 4.2 kb sequence highly similar to the K. pneumoniae

ATCC 13883T chromosomal DNA was identified. This blaSHV-

containing sequence was flanked by IS26 elements in direct orientation.

This supported the notion of a blaSHV-1 transposon in plasmids, however

it was less than half the size estimated by Nugent and Hedges (Nugent

and Hedges 1979). The IS26 element located upstream of blaSHV-1

interrupted lacY, and the downstream IS26 element interrupted the gene

next to blaSHV, ygbI.

The finding that DNA highly similar to the chromosome of K.

pneumoniae surrounded blaSHV in plasmids also added evidence to our

suggestion, in Paper I, that an allelic variant of the chromosomal beta-

lactamase gene in K. pneumoniae is the ancestor of blaSHV genes carried

and spread by plasmids.

Among the plasmid sequences publically available at the time (late 1990s

and early 2000s) only a few contained both blaSHV and IS26, and none of

the sequences contained more than one IS26 element (Naas, Philippon

et al. 1999). The presence of an IS26 element at both ends of the blaSHV-

containing sequence suggested IS26-mediated mobilization of a

chromosomal blaSHV-containing fragment from K pneumoniae to plasmid,

and subsequent mobilization between plasmids. This idea had been

presented earlier, and is now regarded as the most likely evolutionary

scenario (Ford and Avison 2004), (Miriagou, Carattoli et al. 2005),

(Garza-Ramos, Davila et al. 2009). Later, chromosomal origin of other

23

beta-lactamase genes have been demonstrated, e.g., the ancestor of the

currently widespread blaCTX-M genes originates from Klyuvera ascorbata

(Golebiewski, Kern-Zdanowicz et al. 2007). However, these genes are

mobilized by other genetic mobile elements than IS26.

To further analyze the genetic context of blaSHV in plasmids, 11

additional SHV beta-lactamase-producing E. coli strains were included in

the study. These strains were, like E. coli BAB273, from the culture

collection at the section for antimicrobial resistance and infection

control at SMI. They were isolated from patients in different parts of

Sweden 1996 to 2003. There were no epidemiological links between the

patients as judged by information about time and place of isolation. One

of the strains produced a pI 7.6 SHV-1 beta-lactamase and the others

produced different SHV ESBLs conferring resistance, inhibited by

clavulanic acid, to cefotaxime and ceftazidime. Two of the ESBLs were

pI 7.6 enzymes and six were pI 8.2 enzymes.

PCRs developed in order to amplify possible IS26-flanked blaSHV-

containing sequences were positive for all 11 E. coli strains analyzed. In

two strains the sequence had the same length as in E. coli BAB273, in

others it was of different lengths. Sequence comparisons revealed that

all sequences were highly similar to K. pneumoniae chromosomal

DNA. The sequences formed two groups. In each group the flanking

IS26 elements were located at specific distances from blaSHV. In total,

five distinct junctions between IS26 and blaSHV-containing sequence

highly similar to chromosomal K. pneumoniae DNA were identified. Two

of these were located upstream of blaSHV and three downstream of the

beta-lactamase gene.

Later, when the number of sequences publically available had increased

enormously, 33 sequences containing both blaSHV and surrounding

DNA were retrieved from the GenBank nucleotide sequence database

nt/nr by performing a Blastn using our 10.6 kb K. pneumoniae ATCC

13883T sequence as query (Paper III, Table 1). These sequences were

compared to our K. pneumoniae chromosomal and E. coli plasmid

sequences. The comparison demonstrated high diversity among

chromosomal K. pneumoniae sequences, and low diversity among the

plasmid sequences. The published plasmid sequences were from

plasmids carried by different gram-negative hosts, but each of the

sequences was highly similar to one of the two groups identified for the

24

sequences of the 11 E. coli strains analyzed (Paper III, Table 2). Among

the sequences that contained IS26, some had the same IS26-junction(s)

as seen in our E. coli sequences, and some had other junctions. This

resulted in the overall identification of nine distinct IS26-junctions,

which in specific combinations characterize seven lengths of IS26-

flanked blaSHV-containing plasmid sequences (Table 5).

TABLE 5. Characteristics of the seven IS26-flanked blaSHV-containing sequences identified in plasmids carried by different pathogenic gram-negative bacteria

IS26-junctionsa Length (kb) Sequence groupb Plasmid host(s)c

J1 and J9 8.0 PI E. coli, E. cloacae, S. Typhimurium, Y. pestis

J2 and J9 6.7 PI K. pneumoniae

J1 and J6 4.2 PI E. coli

JX and J4 ≥2.4 PI K. pneumoniae

J3 and J8 4.5 PII K. pneumoniae, E. coli, E. cloacae

J3 and J7 1.9 PII E. coli, P. aeruginosa, S. Typhimurium

J3 and J5 1.4 PII E. coli

a Each junction between IS26 and the blaSHV-containing sequence highly similar to K. pneumoniae chromosomal

DNA was numbered on the basis of its position relative to the 10.6 kb fragment of K. pneumoniae ATCC 13883T chromosomal DNA sequence; J1 closest to the 5’-end of the fragment and J9 closest to the 3’-end (Paper III, Fig. 1). Detailed information on which part of the K. pneumoniae chromosome each sequence length corresponds to can be seen in Paper III (Table 1). b The sequence groups are based on SNP analysis of 2.8 kb sequences containing blaSHV, ygbI, and ygbJ

(Paper III, Table 2).

The finding of two plasmid sequence groups was in agreement with data

published later by Ford and Avison (Ford and Avison 2004). They used

bioinformatic approaches to study publically available full-length blaSHV

gene sequences, and constructed an evolutionary tree in which they

identified two main branches. Both branches derived from a blaSHV-1

variant. They concluded that the blaSHV gene had been mobilized at least

twice, and that the mobilization events had been catalyzed by IS26.

In summary, our results support the idea, presented in Paper I, that the

blaSHV variants seen among plasmids originate from the chromosome of

one or more strains closely related to K. pneumoniae ATCC 13883T. They

also support the notion of an evolutionary scenario involving IS26-

mediated mobilization of blaSHV. However, mobilization from

chromosome to plasmid seems to be a rare event. The ongoing spread

of blaSHV genes occurs as IS26-mediated mobilization of blaSHV-

containing DNA fragments, of different lengths, representing two

25

evolutionary lineages. These IS26-flanked blaSHV-containing fragments

are carried by a wide range of plasmids found in many pathogenic gram-

negative bacterial species. Taken together, this implies that

chromosomal K. pneumoniae blaSHV genes are not part of the gene pool

contributing to the current spread of SHV beta-lactamase producing

strains.

26

8 ACKNOWLEDGEMENTS

This has been a long and, in many ways, interesting journey. I have met and worked with many persons during these years, and there are many to thank for help and support. Special thanks go to my supervisors, Sven Löfdahl, Lars G Burman and Barbro Olsson-Liljequist, as well as colleagues in the lab during the years, especially Lasse, Lena, Ingela, Margareta, and Reza.

27

9 REFERENCES

Abraham, E. P. and E. Chain (1940). An enzyme from bacteria able to destroy

penicillin. Nature, 146: 837. Arakawa, Y., M. Ohta, et al. (1986). Close evolutionary relationship between the

chromosomally encoded beta-lactamase gene of Klebsiella pneumoniae and the TEM beta-lactamase gene mediated by R plasmids. FEBS Lett 207: 69-74.

Arpin, C., V. Dubois, et al. (2003). Extended-spectrum beta-lactamase-producing Enterobacteriaceae in community and private health care centers. Antimicrob Agents Chemother 47: 3506-3514.

Babini, G. S. and D. M. Livermore (2000). Antimicrobial resistance amongst Klebsiella spp. collected from intensive care units in Southern and Western Europe in 1997–1998. J Antimicrob Chemother 45:183–9.

Bauernfeind, A., H. Grimm, et al. (1990). A new plasmidic cefotaximase in a clinical isolate of Escherichia coli. Infection 18:294–298.

Bedenic, B. (2002). Selection of Klebsiella pneumoniae mutants with high-level cefotaxime resistance during growth in serum containing therapeutic concentrations of cefotaxime. Chemotherapy 48: 10-14.

Bradford, P. A. (2001). Extended-spectrum beta-lactamases in the 21st century: characterization, epidemiology, and detection of this important resistance threat. Clin Microbiol Rev 14: 933-951.

Brandl, E., M. Giovannini, et al. (1953). Untersuschung über das säurestabile, oral wirksame Phenoxymethylpenicillin (Penicillin V). Wiener Medizinische Wochenschrift 103: 602-7.

Brenner, D. J., J. J. Farmer 3rd, et al. (1977). Taxonomic and nomenclatural changes in Enterobacteriaceae. Atlanta Georgia.

Brisse S., S. Issenhuth-Jeanjean, et al. (2004). Molecular serotyping of Klebsiella species isolates by restriction of the amplified capsular antigen gene cluster. J Clin Microbiol. 42: 3388-3398.

Brisse, S., F. Grimont, et al. (2006). The Genus Klebsiella. Prokaryotes 6:159-196.

Brisse, S. and E. Duijkeren (2005). Identification and antimicrobial susceptibility of 100 Klebsiella animal clinical isolates. Vet Microbiol 105: 307-312.

Brisse, S., C. Fevre, et al. (2009). Virulent Clones of Klebsiella pneumoniae: Identification and Evolutionary Scenario Based on Genomic and Phenotypic Characterization. PLoS ONE 4: e4982.

Brisse, S. and J. Verhoef (2001). Phylogenetic diversity of Klebsiella pneumoniae and Klebsiella oxytoca clinical isolates revealed by randomly amplified polymorphic DNA, gyrA and parC genes sequencing and automated ribotyping." Int J Syst Evol Microbiol 51: 915-924.

Burman, L. G., S. Hæggman, et al. (1992). Epidemiology of plasmid-mediated beta-lactamases in enterobacteria Swedish neonatal wards and relation to antimicrobial therapy. Antimicrob Agents Chemother 36: 989-992.

Bush, K. and G. A. Jacoby (2010). Updated functional classification of beta-lactamases. Antimicrob Agents Chemother 54: 969-976.

28

Bush, K., G. A. Jacoby, et al. (1995). A functional classification scheme for beta-lactamases and its correlation with molecular structure. Antimicrob Agents Chemother 39: 1211-1233.

Datta, N. and P. Kontomichalou (1965). Penicillinase synthesis controlled by infectious R factors in Enterobacteriaceae. Nature 208:239–241.

Edwards, P. R., W. H. Ewing (1972). Identification of Enterobacteriaceae. Burgess Publishing, Minneapolis, MN.

Elhani, D., L. Bakir, et al. (2010). Molecular epidemiology of extended-spectrum beta-lactamase-producing Klebsiella pneumoniae strains in a university hospital in Tunis, Tunisia, 1999-2005. Clin Microbiol Infect 16: 157-164.

Favre-Bonte, S., T. R. Licht, et al (1999) Klebsiella pneumonie capsule expression is necessary for colonisation of large intestine of streptomycin-treated mice. Infect Immun 67: 6152-6156.

Feldman, C., C. Smith, et al. (1990). Klebsiella pneumoniae bacteraemia in an urban general hospital. J Infect 20: 21-31.

Flemming (1929). On the antibacterial action of cultures of a penicillium, with special reference to their use in the isolation of B. influenzæ. Bull. World Health Organ. Vol. 79 no. 8,pp. 780-790, 2001.

Florey, M. E. and H. W. Florey (1943). General and local administration of penicillin. Lancet 1: 387-397.

Ford, P. J. and M. B. Avison (2004). Evolutionary mapping of the SHV beta-lactamase and evidence for two separate IS26-dependent blaSHV mobilization events from the Klebsiella pneumoniae chromosome. J. Antimicrob. Chemother. 54: 69-75.

Fouts, D. E., H. L. Tyler, et al. (2008). Complete genome sequence of the N2-fixing broad host range endophyte Klebsiella pneumoniae 342 and virulence predictions verified in mice. PLoS Genet 4: e1000141.

Francino, M., S. Santos, et al. (2006). Phylogenetic relationships of bacteria with special reference to endosymbionts and enteric species. Prokaryotes 6: 41-59.

Frisch, von, A. (1882). Zur aetiologie des rhinscleroms. Wien. Med Wschr Jahrg 32: 969-972.

Fuster, C., C. Roy, et al. (1993). Frequency of plasmid-mediated beta-lactamases in Enterobacteriaceae isolated in Spain. Eur J Clin Microbiol Infect Dis 12: 67-69.

Galleni, M. and J.-M. Frère (2007). Kinetics of beta-lactamases and penicillin-binding proteins. Enzyme-mediated resistance to antibiotics. R. A. Bonomo and M. Tolmasky. Washington DC, ASM Press: 195-213.

Garcia de la Torre, M., J. Romero-Vivas, et al. (1985). Klebsiella bacteremia: an analysis of 100 episodes. Rev Infect Dis 7: 143-150.

Garza-Ramos, U., G. Davila, et al. (2009). The blaSHV-5 gene is encoded in a compound transposon duplicated in tandem in Enterobacter cloacae. Clin Microbiol Infect 15: 878-880.

Ghuysen, J. M. (1991). Serine beta-lactamases and penicillin-binding proteins. Annu Rev Microbiol 45: 37-67.

Gniadkowski, M. (2008). Evolution of extended-spectrum beta-lactamases by mutation. Clin Microbiol Infect 14: 11-32.

29

Gniadkowski, M., A. Palucha, et al. (1998). Outbreak of ceftazidime-resistant Klebsiella pneumoniae in a pediatric hospital in Warsaw, Poland: clonal spread of the TEM-47 extended-spectrum beta-lactamase (ESBL)-producing strain and transfer of a plasmid carrying the SHV-5-like ESBL-encoding gene. Antimicrob Agents Chemother 42: 3079-3085.

Golebiewski, M., I. Kern-Zdanowicz, et al. (2007). Complete nucleotide sequence of the pCTX-M3 plasmid and its involvement in spread of the extended-spectrum beta-lactamase gene blaCTX-M-3. Antimicrob Agents Chemother 51: 3789-3795.

Grimont PAD, et al. (2005). Genus Klebsiella. In: Brenner DJ, Krieg NR, Staley JT, eds. Bergey’s manual of Systematic Bacteriology Volume 2: The Proteobacteria, Part B: The Gammaproteobacteria. New York: Springer-Verlag. pp 685–693.

Hawkey, P. M. (2008). Molecular epidemiology of clinically significant antibiotic resistance genes. Br J Pharmacol 153 Suppl 1: S406-413.

Hrabak, J., J. Empel, et al. (2009). International clones of Klebsiella pneumoniae and Escherichia coli with extended-spectrum beta-lactamases in a Czech hospital. J Clin Microbiol 47: 3353-3357.

Jacoby, G. A. (1997). Extended-spectrum beta-lactamases and other enzymes providing resistance to oxyimino-beta-lactams. Infect Dis Clin North Am 11: 875-887.

Jacoby, G. A. (2006). Beta-lactamase nomenclature. Antimicrob Agents Chemother 50: 1123-1129.

Jacoby, G. A. and L. S. Munoz-Price (2005). The new beta-lactamases. N Engl J Med 352: 380-391.

Jain, K., K. Radsak (1974). Differentiation of the Oxytocum from Klebsiella by deoxyribonucleic acid-deoxyribonucleic acid hybridization. Int J System Bacteriol 24: 402-407.

Jonas, D., B. Spitzmuller, et al. (2004). Discrimination of Klebsiella pneumoniae and Klebsiella oxytoca phylogenetic groups and other Klebsiella species by use of amplified fragment length polymorphism. Res Microbiol 155: 17-23.

Jones, C. H., M. Tuckman, et al. (2009). Characterization and Sequence Analysis of Extended-Spectrum-beta-Lactamase-Encoding Genes from Escherichia coli, Klebsiella pneumoniae, and Proteus mirabilis Isolates Collected during Tigecycline Phase 3 Clinical Trials. Antimicrob. Agents Chemother. 53 465-475.

Kliebe, C., B. A. Nies, et al. (1985). Evolution of plasmid-coded resistance to broad-spectrum cephalosporins. Antimicrob Agents Chemother 28: 302-307.

Knothe, H. and G. A. Dette (1983). The current state of cephalosporin antibiotics: microbiological aspects. Infection 11 Suppl 1: S12-15.

Knothe, H., P. Shah, et al. (1983). Transferable resistance to cefotaxime, cefoxitin, cefamandole and cefuroxime in clinical isolates of Klebsiella pneumoniae and Serratia marcescens. Infection 11: 315-317.

30

Liu, P. Y., D. Gur, et al. (1992). Survey of the prevalence of beta-lactamases amongst 1000 gram-negative bacilli isolated consecutively at the Royal London Hospital. J Antimicrob Chemother 30: 429-447.

Livermore, D. M. (1991). Mechanisms of resistance to beta-lactam antibiotics. Scand J Infect Dis Suppl 78: 7-16.

Livermore, D. M. (2008). Defining an extended-spectrum beta-lactamase. Clin Microbiol Infect 14 Suppl 1: 3-10.

Livermore, D. M. (2009). beta-Lactamases- the Threat Renews. Curr Protein Pept Sci 10: 397-400.

Livermore, D. M., R. Canton, et al. (2007). CTX-M: changing the face of ESBLs in Europe. J Antimicrob Chemother 59: 165-174.

Livermore, D. M. and M. Yuan (1996). Antibiotic resistance and production of extendedspectrum beta-lactamases amongst Klebsiella spp from intensive care units in Europe. J Antimicrob Chemother 38:409-424.

Lytsy, B., L. Sandegren, et al. (2008). The first major extended-spectrum beta-lactamase outbreak in Scandinavia was caused by clonal spread of a multiresistant Klebsiella pneumoniae producing CTX-M-15. APMIS 116: 302-308.

Massova, I. and S. Mobashery (1998). Kinship and Diversification of Bacterial Penicillin-Binding Proteins and b-Lactamases. Antimicrob Agents Chemother 42: 1–17.

Massova, I. and S. Mobashery (1999). Structural and mechanistic aspects of evolution of beta-lactamases and penicillin-binding proteins. Curr Pharm Des 5: 929-937.

Matthew, M. and A. M. Harris (1976). Identification of beta-lactamases by analytical isoelectric focusing: correlation with bacterial taxonomy. J Gen Microbiol 94: 55-67.

Matthew, M., R. W. Hedges, et al. (1979). Types of beta-lactamase determined by plasmids in gram-negative bacteria. J Bacteriol 138: 657-662.

Mercier, J. and R. C. Levesque (1990). Cloning of SHV-2, OHIO-1, and OXA-6 beta-lactamases and cloning and sequencing of SHV-1 beta-lactamase. Antimicrob Agents Chemother 34: 1577-1583.

Miriagou, V., A. Carattoli, et al. (2005). IS26-Associated In4-type integrons forming multiresistance loci in enterobacterial plasmids. Antimicrob Agents Chemother 49: 3541-3543.

Murray, P. R., K. S. Rosenthal, et al. (2009). Medical microbiology, Elsevier. Naas, T., L. Philippon, et al. (1999). An SHV-derived extended-spectrum beta-

lactamase in Pseudomonas aeruginosa. Antimicrob Agents Chemother 43: 1281-1284.

Nikaido, H. (2003). Molecular basis of bacterial outer membrane permeability revisited. Microbiol Mol Biol Rev 67: 593–656.

Nugent, M. E. and R. W. Hedges (1979). The nature of the genetic determinant for the SHV-1 beta-lactamase. Mol Gen Genet 175: 239-243.

Paterson, D. L., K. M. Hujer, et al. (2003). Extended-spectrum beta-lactamases in Klebsiella pneumoniae bloodstream isolates from seven countries: dominance and widespread prevalence of SHV- and CTX-M-type beta-lactamases. Antimicrob Agents Chemother 47: 3554-3560.

31

Paterson, D. L. and J. G. McCormack (2003). More evidence of the benefits of rational antimicrobial use in clinical practice. Infect Control Hosp Epidemiol 24: 642-643.

Petit, A., H. Ben-Yaghlane-Bouslama, et al. (1992). Characterization of chromosomally encoded penicillinases in clinical isolates of Klebsiella pneumoniae. J Antimicrob Chemother 29: 629-638.

Pitout, J. D. and K. B. Laupland (2008). Extended-spectrum beta-lactamase-producing Enterobacteriaceae: an emerging public-health concern. Lancet Infect Dis 8: 159-166.

Podschun, R. and U. Ullman (1998). Klebsiella spp. as nosocomial pathogens: epidemiology, taxonomy, typing methods, and pathogenicity factors. Clin Microbiol Rev 11: 589-603.

Podschun, R., S. Pietsch, et al. (2001). Incidence of Klebsiella species in surface waters and their expression of virulence factors. Appl Environ Microbiol 67: 3325-3327.

Reading, C. and M. Cole (1977). Clavulanic acid: a beta-lactamase-inhibiting beta-lactam from Streptomyces clavuligerus. Antimicrob Agents Chemother 11: 852-857.

Richards, M. J., J. R. Edwards, et al. (2000). Nosocomial infections in combined medical/surgical intensive care units in the United States. Infect Control Hosp Epidemiol 21: 510-515. Rolinson, G. N. and A. M. Geddes (2007). The 50th anniversary of the discovery of 6-aminopenicillanic acid (6-APA). Int J Antimicrob Agents 29: 3–8.

Rolinson, G. N. and S. Stevens (1961). Microbiological studies on a new broad-spectrum penicillin “Penbritin”. BMJ 2, 191-6.

Sadowski, P. L., B. C. Peterson, et al. (1979). Physical characterization of ten R plasmids obtained from an outbreak of nosocomial Klebsiella pneumoniae infections. Antimicrob Agents Chemother 15: 616-624.

Sauvage, E., F. Kerff, et al. (2008). The penicillin-binding proteins: structure and role in peptidoglycan biosynthesis. FEMS Microbiol Rev 32: 234-258.

Spratt, B. G. (1994). Resistance to antibiotics mediated by target alterations. Science 264: 388-393.

Steward, C. D., J. K. Rasheed, et al. (2001). Characterization of clinical isolates of Klebsiella pneumoniae from 19 laboratories using the National Committee for Clinical Laboratory Standards extended-spectrum beta-lactamase detection methods. J Clin Microbiol 39: 2864-2872.

Tipper, D. J. and J. L. Strominger (1965). Mechanisms of action of penicillins: a proposal based on their structural similarity to acyl-D-alanyl-D-alanine.

Trevisan, V. (1885). Caratteri di alcuni nuovi generi di Batteriaceae. Atti. Accad. fis -med-stat. Milano Ser. 4: 92-106.

Tullus, K., B. Berglund, et al. (1988). Epidemiology of fecal strains of the family Enterobacteriaceae in 22 neonatal wards and influence of antibiotic policy. J Clin Microbiol 26: 1166-1170.

Williams and Thomas (1990). The pathogenicity of Klebsiella pneumoniae. Rev Med Microbiol. 1: 196-204.

Winokur, P. L., R. Canton, et al. (2001). Variations in the prevalence of strains expressing an extended-spectrum beta-lactamase phenotype and

32

characterization of isolates from Europe, the Americas, and the Western Pacific region. Clin Infect Dis 32 Suppl 2: S94-103.

Yigit, H., A. M. Queenan, et al. (2001). Novel carbapenem-hydrolyzing -lactamase, KPC-1, from a carbapenem-resistant strain of Klebsiella pneumoniae. Antimicrob. Agents Chemother. 45:1151–1161.)."

Ørskov, I. (1974). Klebsiella. In: R. E. Bushanan and N. E. Gibbons. eds. Bergey's Manual of determinative bacteriology. Williams and Wilkins, Baltimore, MD. pp. 321-324.